 |
 |
Hello Norman, Yes, You are right, I think something similar is happening for this runs. if you look into the attached figure which represent wall_time wrt file number of 7796 run, for the tpass1, where actually event filter flag was not implemented yet, it shows that all files you mentioned are reconstructed in a significantly less time than the rest. I wanted to check whether svt was moved during that runs through MYA, that would cause less tracks, and consequently small recon but at this moment there is some issue with MYA database for that particular period of time, and I can't retrive that information. Rafo On 07/03/2017 12:35 PM, Graf, Norman A. wrote: > Hello Rafo, > > I noticed that quite a few recon partitions for run 7796 have zero size. > Partitions 230-260 had the SVT problem, so I expected those to be skipped now that we are checking the event flag filter, but was wondering if there was something similar going on for 70, 90, 100, 130, 170. > > Norman > ________________________________________ > From: Rafayel Paremuzyan <[log in to unmask]> > Sent: Friday, June 30, 2017 12:53 PM > To: Graf, Norman A.; [log in to unmask]; hps-software > Subject: Re: test pass status for pass1 > > This is the confluence page for tpass1c > > https://urldefense.proofpoint.com/v2/url?u=https-3A__confluence.slac.stanford.edu_pages_viewpage.action-3FpageId-3D225510613-26flashId-3D-2D1519521923&d=DwIFAw&c=lz9TcOasaINaaC3U7FbMev2lsutwpI4--09aP8Lu18s&r=0HDJrGO9TZQTE97J9Abt2A&m=c9zvHagc18mwEy1jYKlms9IC-VaFaOyumZ21Q4AlWIo&s=LA53665WlDifomowgfOnpTb4Capbanooyav4aayys1A&e= > > let me know, if you have questions/comments > > Rafo > > > On 06/30/2017 03:17 PM, Rafayel Paremuzyan wrote: >> Hello Norman, >> >> This is the jar file, >> /group/hps/hps_soft/hps-java/hps-distribution_vtpass1c_bin.jar >> >> I will create a summary page on the confluence, and will add necessary >> info there. >> >> Rafo >> >> >> On 06/30/2017 03:06 PM, Graf, Norman A. wrote: >>> Hello Rafo, >>> >>> What jar file are you using? >>> >>> Norman >>> ________________________________________ >>> From: [log in to unmask] <[log in to unmask]> >>> on behalf of Rafayel Paremuzyan <[log in to unmask]> >>> Sent: Friday, June 30, 2017 11:56 AM >>> To: [log in to unmask]; hps-software >>> Subject: test pass status for pass1 >>> >>> Hi all, >>> >>> before starting pass1, >>> I submitted a new test pass "tpass1c" with 7796 and 8054 (Carbon target) >>> >>> files are currently being processed, and some of them already have >>> finished >>> >>> Only for this test pass, DQMs were included, but they will not be >>> included during the normal pass, for several reasons >>> - Nobody actually looked at them at the past, and asking about it in >>> several meetings, no one actually requested them to be present >>> - They actually require more memory usage than the recon itself, i.e. >>> rcon is fine with 2.4Gb (Tot Ram from Auger) and 1.4Gb Heap size, while >>> for DQM 1.4 Gb Heap is not enough >>> - DQM takes about 30% of recon time, while other staff (DST and skims) >>> are much faster >>> >>> files are in the following directory >>> /lustre/expphy/work/hallb/hps/data/physrun2016/tpass1c >>> >>> Skims and ntuples in terms of file sizes seems fine to me (though users >>> need to test them). >>> >>> I did simple checks of the event timestamps, it seems ok. >>> >>> So before starting to run the pass1 will be good if users involved in >>> the analysis >>> to check the data. >>> >>> At least following points, which are specifically were addressed for >>> pass1, >>> I think is important to check, and mark as "Good" >>> >>> - Look into skims, make sure it is not too tight, >>> - Check track parameters for detached vertexes >>> - Check whether we get the target position close to 0 >>> >>> If nor issue is found, then probably by the mid of next week I can start >>> pass1. >>> >>> Rafo >>> >>> ######################################################################## >>> Use REPLY-ALL to reply to list >>> >>> To unsubscribe from the HPS-SOFTWARE list, click the following link: >>> https://urldefense.proofpoint.com/v2/url?u=https-3A__listserv.slac.stanford.edu_cgi-2Dbin_wa-3FSUBED1-3DHPS-2DSOFTWARE-26A-3D1&d=DwIFAw&c=lz9TcOasaINaaC3U7FbMev2lsutwpI4--09aP8Lu18s&r=0HDJrGO9TZQTE97J9Abt2A&m=ixwGSS-UD5P_ZfFwhpG0FCSpClbEXFyAztpj-j_UGNg&s=01sS4n3J7bRgQz-hVzhUwx3OzzcP4SH-Ah_CmZbibQg&e= >>> > > ######################################################################## > Use REPLY-ALL to reply to list > > To unsubscribe from the HPS-SOFTWARE list, click the following link: > https://urldefense.proofpoint.com/v2/url?u=https-3A__listserv.slac.stanford.edu_cgi-2Dbin_wa-3FSUBED1-3DHPS-2DSOFTWARE-26A-3D1&d=DwIFAw&c=lz9TcOasaINaaC3U7FbMev2lsutwpI4--09aP8Lu18s&r=0HDJrGO9TZQTE97J9Abt2A&m=c9zvHagc18mwEy1jYKlms9IC-VaFaOyumZ21Q4AlWIo&s=NSTaBIxpeCqyQv1FSUkIYQ90wQjXbfNTUutOX8lYxJo&e= ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the HPS-SOFTWARE list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1